Crystal Structure of Two Anti-Porphyrin Antibodies withPeroxidase ActivityVictor Munoz Robles1., Jean-Didier Marechal1, Amel Bahloul2¤a, Marie-Agnes Sari3, Jean-Pierre Mahy4,

Beatrice Golinelli-Pimpaneau2*.¤b

1 Departament de Quımica, Universitat Autonoma de Barcelona, Edifici C.n., 08193 Cerdanyola del Valles, Barcelona, Spain, 2 Laboratoire d’Enzymologie et Biochimie

structurales, CNRS, Centre de Recherche de Gif, Gif-sur-Yvette, France, 3 Laboratoire de Chimie et Biochimie Pharmacologiques et Toxicologiques, CNRS, Universite Paris

Descartes, Paris, France, 4 Institut de Chimie Moleculaire et des Materiaux d’Orsay, CNRS, Laboratoire de Chimie Biorganique et Bioinorganique, CNRS, Bat 420, Universite

Paris 11, Orsay, France

Abstract

We report the crystal structures at 2.05 and 2.45 A resolution of two antibodies, 13G10 and 14H7, directed against aniron(III)-aaab-carboxyphenylporphyrin, which display some peroxidase activity. Although these two antibodies differ byonly one amino acid in their variable l-light chain and display 86% sequence identity in their variable heavy chain, theircomplementary determining regions (CDR) CDRH1 and CDRH3 adopt very different conformations. The presence of Met orLeu residues at positions preceding residue H101 in CDRH3 in 13G10 and 14H7, respectively, yields to shallow combiningsites pockets with different shapes that are mainly hydrophobic. The hapten and other carboxyphenyl-derivatized iron(III)-porphyrins have been modeled in the active sites of both antibodies using protein ligand docking with the program GOLD.The hapten is maintained in the antibody pockets of 13G10 and 14H7 by a strong network of hydrogen bonds with two orthree carboxylates of the carboxyphenyl substituents of the porphyrin, respectively, as well as numerous stacking and vander Waals interactions with the very hydrophobic CDRH3. However, no amino acid residue was found to chelate the iron.Modeling also allows us to rationalize the recognition of alternative porphyrinic cofactors by the 13G10 and 14H7antibodies and the effect of imidazole binding on the peroxidase activity of the 13G10/porphyrin complexes.

Citation: Munoz Robles V, Marechal J-D, Bahloul A, Sari M-A, Mahy J-P, et al. (2012) Crystal Structure of Two Anti-Porphyrin Antibodies with PeroxidaseActivity. PLoS ONE 7(12): e51128. doi:10.1371/journal.pone.0051128

Editor: Claudine Mayer, Institut Pasteur, France

Received September 6, 2012; Accepted October 30, 2012; Published December 11, 2012

Copyright: � 2012 Munoz Robles et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: Financial support was provided to B.G.-P. and J.-P.M. by the Centre National de la Recherche Scientifique (Programme Physique et Chimie du Vivant)and to J.-D.M. and V.M. by the Spanish "Ministerio de Ciencia e Innovacion" through projects CTQ2008-06866-C02-01 and consolider- ingenio 2010 and by theGeneralitat de Catalunya through project 2009SGR68. The funders had no role in study design, data collection and analysis, decision to publish, or preparation ofthe manuscript.

Competing Interests: The authors have declared that no competing interests exist.

* E-mail: beatrice.golinelli@college-de-france.fr

¤a Current address: Departement de Neuroscience, Institut Pasteur, Paris, France¤b Current address: Laboratoire de Chimie des Processus Biologiques, College de France, CNRS, Paris, France

. These authors contributed equally to this work.

Introduction

Hemoproteins contain iron-protoporphyrin IX or heme as the

prosthetic group, whose divalent iron atom can reversibly bind

molecules such as molecular oxygen, leading to a wide range of

biological functions [1]. Chemical or biotechnological models of

hemoproteins have thus long been developed in order to create

selective catalysts for industrial and fine chemistry and to predict

the oxidative metabolism of new drugs [2,3,4,5]. Examples include

the de novo design of heme proteins, including that of membrane-

soluble proteins [6,7]. Peroxidases appear to be the easiest

hemoproteins to be mimicked. Indeed, their active site consists

of the iron(III)-porphyrin moiety encapsulated in the apoprotein.

On one side, the heme iron is bound to an axial histidine residue

(proximal ligand) and on the other side to the peroxide substrate to

lead to an iron-oxo complex. The radical cation on the iron (IV)-

oxo porphyrin ring can be delocalized onto proximal protein side

chains [8]. The reducing cosubstrate does not bind to a well-

defined site on the inside of the protein, as peroxidases restrict

access of substrates to the heme-oxo complex, so that the electron

transfer occurs to the meso edge of the heme [9]. Heterolytic

cleavage of the O-O bond is assisted by general acid base catalysis

through the concerted action of the distal histidine and arginine

residues [10]. A major problem in homogeneous metalloporphyrin

systems mimicking hemoproteins is that the catalyst is often

destroyed by oxidation during the course of the reaction and it is

difficult to combine reactivity and selectivity in these models. The

use of a protein such as xylanase A [11] or an antibody mimicking

the protein matrix of heme enzymes not only prevents aggregation

and intermolecular self-oxidation of the catalyst, but can also

influence the selectivity of the reaction [12]. As the antibody has

the role of a host molecule that enhances the function of

porphyrin, the porphyrin itself can be used as the hapten to

induce the antibodies.

In order to generate antibodies with peroxidase activity, mice

have been immunized against iron(III)-a,a,a,b-meso-tetrakis-ortho-

carboxyphenyl-porphyrin (Fe(ToCPP)) (Figure 1) [13,14]. Two

antibodies, 13G10 and 14H7, were found to bind the porphyrin

hapten with nanomolar affinities and enhance its peroxidase

PLOS ONE | www.plosone.org 1 December 2012 | Volume 7 | Issue 12 | e51128

activity. The 13G10-Fe(ToCPP) and 14H7-Fe(ToCPP) complexes

catalyzed the oxidation of 2,29-azinobis (3-ethylbenzothiazoline-6-

sulfonic acid) (ABTS) 5 to 8 fold more effectively than the cofactor

alone, with kcat reaching 540 min21 and kcat/Km(ABTS) 6.2

104 M21 min21 at pH 4.6 for the 13G10 complex [15]. The

antibodies protected the cofactor from oxidative degradation,

allowing more than one thousand turnovers before destruction. In

addition, it was shown that the 13G10-Fe(ToCPP) complex

possessed a remarkably thermostable peroxidase activity and that

it was able to use not only H2O2 as the oxidant but also a wide

range of hydroperoxides [15,16]. Compared with the peroxidase

enzymes, the two antibodies possess a low affinity toward H2O2

(Km = 9–16 mM) but a higher affinity than the cofactor alone

(Km = 42 mM) [13]. Because no amino acid was found to

coordinate the iron atom in the 13G10 and 14H7/Fe(ToCPP)

complexes [14], it was thought that an imidazole ligand could

mimic the proximal histidine of peroxidases and enhance the

catalytic activity. Fe(ToCPP) and less hindered iron(III)-tetraar-

ylporphyrins, bearing only one or two carboxyphenyl substituents,

were found to bind only one imidazole ligand when complexed to

antibody 13G10, suggesting that the complexes are good mimics

for peroxidases [16]. Whereas 50 mM imidazole inhibited the

peroxidase activity of the 13G10-Fe(ToCPP) complex, it increased

that of 13G10 in complex with porphyrins bearing two meso-

[ortho-carboxyphenyl] substituents by a factor 15.

To get insight into the structural basis for the different binding/

activities of the variously substituted porphyrins, we have

determined the crystal structure of the fragment antigen binding

(Fab) of antibodies 13G10 and 14H7 and modeled their

interaction with the different iron-porphyrins using molecular

docking.

Materials and Methods

Purification and Characterization of Fabs 13G10 and14H7

The antibodies were produced in ascitic fluid as described [13].

After ammonium sulfate precipitation, the IgGs (IgG1, l) were

loaded on a protein A column in 3 M NaCl, 1.5 M glycine pH 8.9

and eluted by 0.1 M citrate pH 5. The Fab 13G10 (resp. 14H7)

was generated by papain digestion of the antibody at 37uC under

standard conditions (30 mM Tris pH 7.4, 138 mM NaCl,

1.25 mM EDTA, 1.5 mM 2-mercaptoethanol) using a 3% papain

to antibody ratio (w/w) and a 10 h (resp. 8 h) digestion time.

Undigested IgG and Fc fragment were removed by DEAE anion

exchange chromatography followed by gel filtration on a

Sephacryl S100 HR column. The Fabs were further purified by

ion exchange chromatography on a mono Q FPLC column by a

NaCl gradient in 20 mM bistrispropane buffer at pH 7.2 for Fab

14H7 or in 20 mM diethanolamine buffer at pH 8.8 for Fab

13G10.

Sequence DeterminationN-terminal sequencing of the H chain of 13G10 showed that the

first three amino acid residues of the Fab were missing. mRNA

from 13G10 hybridoma cells (4 106 cells) were isolated using a

mRNA purification kit from Dynal. cDNAs were prepared in 50ml

using the Omniscript Reverse Transcriptase kit from Qiagen and

stored at 220uC until use. The heavy chain variable region

fragment (about 420 bp) of each cDNA was amplified using a set

of 12 forward primers MHV1 to MHV12 previously described

[17] and reverse primer IgG1: 59GGATCCCGGGCCAGTG-

GATAGACAGATG complementary to the beginning of the

constant region. The l-light chain fragment (about 690 bp) of

each cDNA was amplified using a mixture of 2 forward primers,

NL1: 59ATGGCCTGGATTTCACTTATAC and NL2:

59ATGGCCTGGACTCCTCTCTTC, corresponding to mouse

l-light chain leader sequence, and a mixture of 4 reverse primers,

CL1 59GCAGGAGACAGACTCTTCTCCAC, CL2 59GCAC-

GAGACAGACTCTTCTCCAC, CL3 59GCAGGGGA-

CAAACTCTTCTCCAC, and CL4 59GCACGGGA-

CAAACTCTTCTCCAC, complementary to mouse CLterminal

sequence.

Each cDNA mixture (3 ml) was amplified with the Polymerase

Chain Reaction [18] using TFL polymerase (Promega) on a MJ

Research mini cycler with the following programm. A 10 min

denaturation step at 94uC was followed by 30 cycles of 30 sec

denaturation at 94uC, 45 sec hybridization at 59uC and 1 min

30 sec elongation at 72uC, a final step of 10 min at 72uC was

performed to ensure completion of the amplification. Amplified

products were purified on 0.8% low melting agarose (Sigma) and

ligated into pGEMHTeasy vector (Promega). Electrocompetent E.

coli TG1 cells {D(lac pro) supE thi hsdD5 F’ traD35 proAB LacIq

LacZDM15} were transformed with the ligation mixture by

electroporation using a Cell porator electroporation system

equipped with a Voltage Booster (Life technologies) according to

manufacturer’s recommendations. Plasmid DNAs were extracted

from transformed cells and submitted to dideoxy sequencing [19].

For each antibody, clones were originated from at least two

independant Polymerase Chain Reactions.

CrystallizationCrystals of 13G10 were grown in 26.5% PEG 2000, 0.2 M

MgCl2, 0.1 M Tris pH 8.5, 10% glycerol. Crystals were flash

cooled in a nitrogen stream at 100 K in the same solution

containing 10% glycerol. Crystals of 14H7 were grown in 25%

PEG 4000, 0.1 M ammonium acetate, 0.1 M sodium cacodylate

pH 6.5. A single capillary-mounted crystal kept at 4uC was used

for data collection. This explains the low redundancy for the 14H7

data.

X-Ray data collection and structure

determination. Diffraction data for Fab 13G10 and Fab

14H7 were recorded at the ID14-1 station of ESRF and the

Figure 1. General structures of the cofactors used in this work.R1a = R2a = R3a = R4b = COOH, R2b = H: iron(III)- a,a,a,b-mesotetrakis-orthocarboxyphenyl-porphyrin Fe(ToCPP) hapten used to induce theantibodies 13G10 and 14H7; R1a = R2a = COOH, R2b = R3a = R4b = H: aa-Fe(DoCPP); R1a = R2b = COOH, R2a = R3a = R4b = H: ab-Fe(DoCPP);R1a = COOH, R2a = R2b = R3a = R4b = H: Fe(MoCPP).doi:10.1371/journal.pone.0051128.g001

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 2 December 2012 | Volume 7 | Issue 12 | e51128

LURE DW32 station, respectively. Data were processed with

DENZO and SCALEPACK [20] (Table 1). The structures were

solved by molecular replacement with the program AMoRe [21];

the models used were the Fv domain (PDB code 1mfa) and the

CL-CH1 dimer (PDB code 1mfe) of the murine anticarbohydrate

antibody Se155-4, which belongs to the same IgG1, l class [22].

The atomic model of 13G10 was refined alternating cycles of

model reconstruction with O [23] and refinement with CNS [24],

whereas the atomic model of 14H7 was refined alternating cycles

of model reconstruction with COOT and refinement with PHENIX

[25] using the twin option. The final refinement statistics are given

in Table 1. Several residues at the N-terminus of the heavy chain

and CDRH1 of four heavy chains that were disordered were not

included in the model. Figures were drawn with PYMOL. ThrL51

in CDRL2, which lies in the disallowed region of the Ramachan-

dran plot, belongs to a c turn, as commonly observed in all

antibody structures.

Molecular ModelingQuantum mechanical calculations were carried out to model

the structures of the tetra-, bi- and mono substituted porphyrins.

These structures were fully optimized with the density functional

B3LYP [26,27], as implemented in Gaussian09 [28]. The double-z

basis set LANL2DZ [29] and its associated pseudo potential were

used for the iron and the split valence 6–31 g** [30,31] for the

other atoms (C, N, H and O). Calculations were carried out for the

high spin ferric species with the iron coordinated by the four

porphyrin nitrogen atoms. The structures of the 13G10 and 14H7

antibodies were processed with the UCSF Chimera package [32]

prior to docking in order to remove the water, ions and glycerol

molecules, calculate the protonation states of the amino acids and

add hydrogen atoms. Protein-ligand dockings were undertaken for

both antibody structures following a recently published protocol

that accounts for the presence of the metal in the ligand [33].

Calculations were carried out using the program GOLD (version

Table 1. Data collection and refinement statistics for Fab 13G10 and 14H7.

Data collection Fab13G10 Fab14H7

Space group C2 P21

Number of molecules in the asymmetric unit 1 8

unit cell a = 55.8 A, b = 62.7 A, c = 113.5 A a = 56.2 A, b = 228.2 A, c = 146.6 A

a= 90, b= 91.7u, c= 90u a= 90, b= 90.1u, c= 90u

resolution 25–2.05 A 20–2.55 A

(outer resolution shell) 2.09–2.05 A 2.64–2.55 A

unique reflections 23580 111576

completeness* 93.7% (91.4%) 93.7% (87.8%)

mean I/sigma* 23.7 (4.6) 7.4 (2.1)

Rsym* 0.037 (0.19) 0.109 (0.51)

redundancy* 2.5 (2.1) 1.2 (1.2)

Refinement statistics

resolution 20.0–2.05 20.0–2.45

protein residues 425 3356

water molecules 188 359

glycerol molecules 3 –

Mg2+ molecule 1 –

Rcryst 0.224 0.279

Rfree{ 0.275 0.328

Deviations from ideal geometry (rms)

bond length deviation (A) 0.008 0.003

bond angle deviation (u) 1.52 0.75

B values

Average B value of protein atoms (A2) 49.0 46.5

Average B value of water molecules (A2) 51.1 25.5

Average B value of glycerol (A2) 76.3 –

B value of Mg2+ (A2) 45.0 –

Ramachandran plot

Most favored (%) 86.1 79.6

Additionally allowed (%) 11.4 17.5

Generously allowed (%) 1.1 1.4

Disallowed (%) 1.4 1.5

*Values for highest-resolution shell are given in parentheses.{4.5% and 4.7% of the data were set aside for the Rfree calculation during the entire refinement.doi:10.1371/journal.pone.0051128.t001

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 3 December 2012 | Volume 7 | Issue 12 | e51128

5.1) [34] and the Chemscore [35] scoring function. The metal ion

was modeled thanks to an in house parameterization of an

additional atom type in GOLD, which mimics the chelating

capacity of the metal by interacting with amino acids with

potential Lewis basis capacities. For the docking of imidazole, the

parameters of the Chemscore scoring function used for the iron

atom were that of cytochrome P450, as implemented in

GOLD5.0. Because the Mg2+ ion lies at the interface between

the hypervariable loops in the structure of 13G10 and probably

mimics the iron of the porphyrin hapten, the docking calculations

were performed using a 20 A sphere around the Nd atom of the

neighboring residue AsnH33. Based on initial rigid dockings, full

flexibility was allowed for TyrL34 using the Dunbrack rotamer

library [36]. The bonds between the carboxyphenyl groups and

the porphyrin moiety were frozen to maintain the correct a/bconfiguration. For each docking, an in house approach, based on a

statistical analysis of a large PDB set of metal-bound proteins, was

used in order to identify those residues that could directly bind to

the metal. Based on the distances between the metal and the Cacarbons of all neighboring residues, no direct interaction was

identified between the metal and the protein. Therefore, no

further QM/MM refinements of the complexes were carried out

[33].

Results

Amino Acid Sequence of Fab 13G10 and Fab 14H713G10 and 14H7 are murine monoclonal antibodies that were

found to belong to the IgG1, l-class. The sequences of their

variable chains (Figure 2) differ by one residue in the light chain,

outside of the complementary determining regions (CDRs) [37],

and 17 residues in the heavy chain. In CDRH1, the sequences

differ at positions H23 (Lys/Thr), H28 (Thr/Ser), H30 (Thr/Ser)

and H34 (Met/Ile) for 13G10 and 14H7, respectively. CDRH2

has the same sequence in both antibodies. Their CDRH3, which is

a main determinant of the combining site, has the same length but

contains, in addition to Ser/Ala mutations at positions 93 and 97,

two different amino acids with the same hydrophobic nature at

position H100b (Ile/Leu) and H100c (Met/Leu).

Mature antibody genes are first formed by the assembly of the

variable (V) and junction (J) gene segments for the light chain

variable (VL) domain and by the assembly of the variable (V),

junction (J) and diversity (D) gene segments for the heavy chain

variable (VH) domain; affinity maturation then introduces into

antibodies somatic mutations that increase binding affinity to the

hapten or antigen [38]. For mice (Mus Musculus) antibodies, only

three germline Vlsegments (compared with 180 Vk segments),

and four Jl segments have been uncovered [39], which explains

the poor diversity of the l-light chain variable sequences

compared with the k-class. Compared with the germline Vl1/

Jl1 sequence [40], there is one difference in the variable light

chain of 13G10 and 14H7, Leu at position L96 (instead of Trp) in

CDRL3 that arises from diversity at the junction between the V

and J gene segments (Figure 2). In addition, 14H7 displays a Thr

instead of a Lys at position L39, which is located outside the

recombinant site. The nucleotide sequence of the 13G10 and

14H7 VH domains shows that they are derived from a gene in the

J558 family, the largest VH family. The 13G10 VH sequence has

highest homology to germline gene sequence J558.42 (98%

identity at the nucleotide level) [41]. This translates into 94 (resp.

87) identical amino acid residues out 98 in VH for 13G10 and

14H7, respectively (Figure 2). The nucleotide sequence indicates

that the VH region of antibodies 13G10 and 14H7 is encoded by

the combination of J558.42 with a D gene segment that could not

be identified because it is too short and JH4.

Comparison of the 13G10 and 14H7 VL and VH sequences to

that of the germline genes indicates that 13G10 and 14H7 did not

undergo much affinity maturation in the V segments. Among the

rare somatic mutations, the ones that belong to the combining sites

are LeuL96 (CDRL3), ValH50 (CDRH2) and SerH93 (CDRH3),

for 13G10 (Table 2, Figure 2). On the other hand, the exceedingly

hydrophobic nature of CDRH3 of 13G10 and 14H7 is due to

extensive affinity maturation in the D segment so that the germline

D segment could not be identified from the nucleotide sequence.

Alanine at position H101 comes from a somatic mutation.

Determination of the Structures of Fab 13G10 and Fab14H7

The crystal structures of Fab 13G10 and Fab 14H7 were

determined at 2.05 A and 2.55 A, respectively (Table 1). Fab

13G10 crystallized with one molecule in the asymmetric unit,

whereas the crystals of Fab 14H7 contained eight Fab molecules in

the asymmetric unit and were pseudomerohedrally twinned

[42,43,44]. Twinning was detected using the xtriage program

included in PHENIX [25]. The cumulative local intensity deviation

distribution statistics [45] did not indicate twinning:

,|L|. = 0.488 (untwinned: 0.500; perfect twin: 0.375);

,L2. = 0.319 (untwinned: 0.333; perfect twin: 0.200). However,

the results of the H-test [46] on acentric data gave a ,|H|. value

of 0.046 (0.50: untwinned; 0.0:50% twinned) and a ,H2. value

of 0.005 (0.33: untwinned; 0.0:50% twinned). The estimation of

the twin fraction by the H-test was 0.454 compared with 0.425 by

Britton analysis [47] (Figure S1). The twinning law h, -k, -l and a

twinning fraction of 0.5 were used throughout the refinement

using PHENIX.

Overview of the Structures of Fab 13G10 and Fab 14H7Fab 13G10 and Fab 14H7 display large elbow angles of 175.3u

and 170.1u, respectively, in agreement with their light chain

belonging to the l-class. The frameworks of the two Fabs

superimpose with an rmsd of 1.067 A2 over 174 Ca atoms.

However, the fit is better for the VL and VH domains taken

separately, with an rmsd of 0.54 A2 over 89 Ca atoms and 0.92 A2

over 85 Ca atoms, respectively. This reflects a difference in the

orientation of the VH and VL domains of the two antibodies with

respect to each other of 8.04u and 0.61 A. Together with the

different conformations of CDRH3 (see below), the combining site

of 14H7 is more open, with an accessible surface area of 19810 A2

compared with 18820 A2 for 13G10 (Figure 3A).

The hapten-binding pocket of 13G10 is a shallow cleft, roughly

12.7 A by 7.2 A wide and 8 A deep, at the upper part of the VL/

VH interface (Figure 3B and 3C). Electron density could be

observed for a magnesium ion and two molecules of glycerol in the

combining site of Fab 13G10 (Figures 3B and 4A). The

magnesium ion likely indicates the position of the Fe(ToCPP)

iron in the hapten/antibody complex, thus confirming the location

of the catalytic pocket at the antibody combining site at the VL/

VH interface.

In 13G10, CDRH1 adopts the canonical structure H1-13-1

(Figure 4A) [37]. In 14H7, electron density is observed for

CDRH1 only for four molecules in the asymmetric unit out of

eight and it adopts a very different conformation from that in

13G10, which has not yet been listed (Figure 4B) [37]. This is a

consequence of the four differences in amino acids in this CDR

between the two antibodies. However, this difference in the Cabackbone of CDRH1 does not contribute to create different

topologies in the combining sites because only AsnH33, whose side

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 4 December 2012 | Volume 7 | Issue 12 | e51128

Figure 2. Amino acid sequences of the VL and VH domains of 13G10 and 14H7. Dots denote sequence identity to antibody 13G10. Hdenotes the heavy chain and L the light chain. Numbering of the antibody residues follows the Kabat nomenclature [40] and the definition of thehypervariable regions, indicated in bold, is from North et al. [37]. The sequence in italics has also been determined by Edman degradation of theaminoterminal part of the protein. The nucleotide sequences of 13G10 and 14H7 have been deposited in the Genbank database, except for the

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 5 December 2012 | Volume 7 | Issue 12 | e51128

chain has two different conformations in the two antibodies,

together with HisH35, line the pocket (Figure 3C). CDRH2 adopts

the canonical structures H2-10-1 in both antibodies [37] but the

side chains of TyrH52 display very different conformations

(Figure 4C). CDRH3 in both antibodies does not share the same

conformation because of the flexibility of the main chain due to

the presence of three glycine residues (Figure 2). AsnH33, TyrH52

and the main chain of CDRH3 are in contact so that their

conformation depends on each other (Figure 4C). In both 13G10

and 14H7, CDRL1, CDRL2 and CDRL3 adopt the canonical

structures L1-14-1, L2-8-4 and L3-9-1, respectively [37]. The side

chain of TrpL91 in CDRL3 has different orientations in the two

antibodies (Figure 4C).

AsnH33, several residues of CDRH2 (TyrH52, AspH56,

SerH58 and the somatically mutated residue ValH50) and

CDRH3 form one face of the combining site (Figure 3B and

3C). The main residues lining the other face of the cavity are

TyrL32 and TrpL91. The bottom of the site is composed of

HisH35 and somatically mutated LeuL96. The identity of the

amino acids that define the combining site of 13G10 and 14H7 is

consistent with their belonging to the l-light chain class of

antibodies (Table 2). The very hydrophobic nature of the antigen-

binding site is not only due to the presence of Tyr, Trp and Leu

constant heavy chains whose sequences have not been determined: Genbank accession numbers 13G10H, AY178830; 13G10L, AY178831 for themRNA and AA020092, AA020093 for the corresponding protein sequence; 14H7H, AY178829; 14H7L, AY178828 for the mRNA and AA020092,AA020093 for the corresponding protein sequence. The previously published amino acid sequences [14] were not those of antibodies 13G10 and14H7. The germline sequences are indicated below the sequence of the antibodies.doi:10.1371/journal.pone.0051128.g002

Table 2. Comparison of the aminoacids that contact small haptens in the structurally characterized l-light chain antibodies.

PDB code13G10/14H7*

Se155-41mfa [22]

CHA2551ind [79]

NC101etz [83]

88C6/121yuh [80]

2D12.51nc2 [85]

RS2-G192ntf [82]

10G61jnh [84]

N1G91ngp [81]

L32 CDRL1 Tyr His Tyr Tyr Tyr Tyr Tyr Tyr Tyr

L34 – Asn Asn Asn Ile Asn Asn Asn Asn Asn

L91 CDRL3 Trp Trp Trp Trp Trp Trp Trp Trp Trp

L93 – Ser Asn Ser Ser Ser Ser Ser Ser Ser

L94 – Asn Asn Asn Asn Asn Asn Asn Asn Asn

L96 – Leu Trp Trp Trp Trp Trp Trp Phe Trp

H33 CDRH1 Asn Trp Thr Gly Leu Gly Trp Trp Trp

H35 – His His Ser Gly His His His Gln His

H47 FR2 Trp Trp Trp Leu Trp Trp Trp Trp Trp

H50 CDRH2 Val Ala Thr Asp Arg Val Thr Ala Arg

H52 – Tyr Tyr Leu Trp Asp Trp Tyr Tyr Asp

H52A – Pro Pro Ser Pro Pro Pro Pro

H53 – Gly Asn Gly Asn Asn Ser Gly Gly Asn

H56 – Asp Ala Phe Lys Val Gly Asn Asp Gly

H58 – Ser Phe Phe Tyr Lys Ala Tyr Arg Lys

H95 CDRH3 Gly Gly His Arg Tyr Arg Gly Gly Tyr

H96 – Gly Gly Arg Thr Ala Gly Ser Arg Asp

H97 – Ala/Ser His Phe Tyr Ser Leu Ser Tyr

H98 – Ser Cys Tyr Tyr Leu Tyr

H99 – Tyr Arg Pro Tyr Tyr Gly

H100 – Tyr Tyr Asn Ser

H100 – Tyr

H100 – Gly

H100 – Ser

H100 – Ser Asn

H100 – Gly Phe Tyr

H100 – Gly Tyr Tyr Asn Gly Tyr Ser

H100 – Ile Tyr Tyr Pro Tyr Trp Thr Tyr

H100 Met/Leu Gly Phe Met Phe Phe Met Phe

H101 Ala Asp Val Asp Asp Asp Gly Asp Asp

Underligned letters are residues that form direct or water-mediated hydrogen-bonds or salt-bridge to the hapten. A bold letter indicates that the residue is in van derWaals contacts with the hapten.*For 13G10 and 14H7, the hapten has been modeled by docking (see Table 3).doi:10.1371/journal.pone.0051128.t002

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 6 December 2012 | Volume 7 | Issue 12 | e51128

residues at contact positions, as observed in other antibody

structures (Table 2). In addition, the CDRH3 loops of 13G10 and

14H7 are very peculiar in that they contain a high number of

glycine and alanine residues (H95-H100a) and only a few residues

with a long side-chain that could interact with the hapten

(Figure 2).

Figure 3. General view of the combining sites of Fab 13G10 and Fab 14H7. A Comparison of the binding site cavities of 13G10 and 14H7.The VL frameworks of 13G10 (in blue) and 14H7 (in green) have been superimposed. The location of Mg2+ in 13G10 is indicated as a blue ball. BGeneral view of the active site of 13G10. CDRL1, CDRL2, CDRL3, CDRH1, CDRH2 and CDRH3 are shown as red, pink, orange, magenta, green and cyanribbons, respectively. Mg2+ and two glycerol molecules are indicated as blue ball and sticks, respectively. C Zoom of the combining site of 13G10.doi:10.1371/journal.pone.0051128.g003

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 7 December 2012 | Volume 7 | Issue 12 | e51128

Molecular Modeling of the a, a, a, b Hapten-antibodyComplexes

Because we did not succeed to grow crystals of Fab 13G10 and

Fab 14H7 in complex with the Fe(ToCPP) hapten, either by co-

crystallization or soaking, Fe(ToCPP) was modeled in the

recombinant site by molecular docking using the Gold 5.0 package

[34] and a previously reported protocol [33] (Figures 5A, 5E,

6A and 6E; Table 3). The docking solutions display the adjacent

a1 and a2 carboxyphenyl substituents deeply buried inside the

binding site, while those in a3 and b configuration are exposed

toward the solvent. The comparison of the accessible surface areas

of Fe(ToCPP) alone and in complex with 13G10 and 14H7

indicates that 53.7% and 44.9%, respectively, of the porphyrin is

buried in the antibody pocket.

In the most stable 13G10/Fe(ToCPP) predicted complex, the

hydrophobic CDRH3 loop stacks against the most buried

carboxyphenyl group of the porphyrin hapten (Figure 5A) and

must therefore have been specifically selected by the immune

system. Ligand recognition is achieved through H-bonds

(Figure 5A), van der Waals contacts and stacking interaction

(Table 3 and Figure 6A). In the calculated complex, the metal of

the porphyrin is located close to the magnesium atom that has

been characterized in the X-ray structure (Figure 3B and 4A,

Table 3). In the 20 first solutions with low energy for the 14H7/

Fe(ToCPP) complex, the porphyrin is always located in the

solvent-exposed region of the binding site (Figure 5E and 6E). This

leads to very variable orientations of the cofactor with weaker

complementarity to the antibody than in 13G10. The best solution

obtained for 14H7 is 10 kJ/mol less stable than that for 13G10

because of more extensive hydrophobic interactions in the case of

13G10 (Table 3). Indeed, while the hydrogen bonding energy is

the same for the two complexes, the lipophilic energy is about

Figure 4. Comparison of the CDRs of Fab 13G10 and Fab 14H7. A Peculiar conformation of CDRH1 in the combining site of Fab 13G10.CDRH1 is shown as magenta sticks and a 2Fo-Fcalc map, contoured at 1 s and displayed around CDRH1, Mg2+ and the two glycerol molecules, as agrey mesh. B Comparison of the conformations of CDRH1 in 13G10 and 14H7. CDRH1 in 14H7 and 13G10 are shown as magenta sticks and blacklines, respectively. The VH frameweworks of the two antibodies were superimposed. A 2Fo-Fcalc map contoured at 1 s is displayed around CDRH1 of14H7 and shown as a grey mesh. C Comparison of the combining sites of Fab 13G10 (CDRs colored as in Fig. 4A) and Fab 14H7 (black).doi:10.1371/journal.pone.0051128.g004

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 8 December 2012 | Volume 7 | Issue 12 | e51128

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 9 December 2012 | Volume 7 | Issue 12 | e51128

100 kJ/mol lower in 13G10. Moreover, less clashes can be

observed in 14H7, compared with 13G10. These observations are

consistent with a lesser degree of complementarity between the

cofactor and the antibody in 14H7 than in 13G10. In the former,

the cofactor remains solvent exposed, with the most important

interactions taking place with TyrH32, TyrH52 and SerH97,

while in the latter, the cofactor binds more deeply inside the cavity

with the H-bonding interactions taking place with more buried

residues: AsnH33, HisH35 and the NH groups of GlyH96 and

GlyH100a.

Our models of the complexes of 13G10 and 14H7 with

Fe(ToCPP) indicate that no residue in the vicinity of the cofactor is

able to chelate the metal in either antibody. The nearest residue

(AsnH33 in 13G10 and TyrH32 in 14H7) has its Nd or OH atom

4 A and 4.5 A away, respectively, from the metal. Apparently, the

steric hindrance caused by the carboxyphenyl groups prevents the

iron from approaching any residue of the protein.

Rationalization of the Binding of Alternative CofactorsTo obtain better mimics of peroxidases, capable of binding one

imidazole ligand, complexes of 13G10 and 14H7 with less

hindered iron(III)-tetraarylporphyrins, bearing only one or two

carboxyphenyl substituents (monosubstituted Fe(MoCPP) and a,bor a,a, di-substituted Fe(DoCPP)), were designed (Figure 1) [16].

Although all 13G10/porphyrin complexes were shown to bind

only one imidazole ligand, the affinity for imidazole of the a,a and

a,b-Fe(DoCPP) complexes was 2–3 fold lower than that of

13G10/Fe(ToCPP). Interestingly, the 13G10/dicarboxyphenyl

porphyrin complexes presented higher activity for ABTS oxidation

than the original Fe(ToCPP)/13G10 complex in the presence of

50 mM imidazole. The crystal structures provided in this work

combined with molecular modeling offer the opportunity to

rationalize such findings.

First, the mono and disubstituted porphyrins were docked into

the combining sites of the two antibodies (Figure 5B–D, 6B–D,

7A and B, Table 3). Overall, the four porphyrins have relatively

similar binding modes to 13G10, with a good complementarity of

a major part of the macrocycle with the binding site (Figure 6A–D

and 7A). A strong interaction between one of the porphyrin

carboxylates and both AsnH33 and HisH35 of CDRH1 is always

present and appears as the most important feature in the

recognition of the cofactors by 13G10. This is sustained by the

fact that the complexes with Fe(ToCPP) and Fe(MoCPP), despite

differing by three carboxylates and making five and three H-bonds

with 13G10, respectively, have similar binding energies (Table 3).

Thus, increasing the number of carboxylates in the cofactor does

not mean an overall better binding because it also increases steric

hindrance and results in more important bad contacts (Table 3).

Slight differences are still observed in the way the ligands penetrate

into the cavity, with aa-Fe(DoCPP) displaying the highest buried

surface, the lowest one being obtained for Fe(ToCPP).

Fe(MoCPP) binds to 14H7 in a very different way compared

with Fe(ToCPP) (Figure 7B). For ab-Fe(DoCPP), the binding

mode to 14H7 is very similar to that of Fe(ToCPP), while for aa-

Fe(DoCPP), the porphyrin is noticeably displaced. However,

overall, the binding modes of the three alternative cofactors are

similar in energy, including for its individual terms (Table 3).

Docking imidazole in the various porphyrin/13G10 complexes

indicates that two molecules of imidazole can bind the iron, one on

each side of the cofactor. However, imidazole binds preferentially

on opposite faces of the porphyrin, depending on the cofactor

(Figure 7C, D and E). For Fe(ToCPP), imidazole binds to the

solvent-exposed face, whereas in the case of the dicarboxylates-

containing porphyrins, it binds in the cavity formed between the

cofactor and the protein. The overall binding energies are very

similar and the main differences appear in the hydrogen bond and

lipophilic terms (Table 4).

Structural Basis of Catalysis: Comparison to theFerrochelatase and Porphyrin-dependent PeroxidaseAntibody 7G12

The structure of 13G10 can be compared with that of the k-

light chain metallochelatase and porphyrin-dependent peroxidase

antibody 7G12 (Fig. 8A and B). Antibody 7G12, induced against a

distorted N-methylmesoporphyrin IX, a transition state analogue

for porphyrin metalation [48], catalyzes the chelation of various

metals by mesoporphyrin IX. In addition, the complex of 7G12

with Fe(III)-mesoporphyrin IX was shown to catalyze the

oxidation of several chromogenic peroxidase substrates by

hydrogen peroxide with more than 200 turnovers with kcat of

6 s21 and kcat/Km of 300 M21 s21 [49]. Thus, the peroxidase

activity of 7G12 is similar to that of 13G10. The crystal structure

of antibody 7G12 complexed with N-methylmesoporphyrin IX has

shown that the antibody induces geometric strain in the porphyrin

substrate to catalyze porphyrin metalation [50,51,52]. The

carboxylate side-chain of AspH96 of 7G12, which is positioned

1.9 A from the center of the porphyrin ring (Fig. 8A), is thought to

act as a catalytic residue in the metal chelatase reaction by

Figure 5. Models of the complexes of 13G10 and 14H7 with the different porphyrins. The H-bonds between the antibodies and theporphyrins are indicated. A 13G10/Fe(ToCPP). Molecular docking gives two most favorable orientations of Fe(ToCPP) bound to 13G10. In the moststable 13G10/Fe(ToCPP) predicted complex, the carboxylate of the b substituent makes a H-bond with SerH58, while the a1 substituent is H-bondedto AsnH33, HisH35, GlyH96 and GlyH100a. Only two bad contacts between the ligand and protein are observed in this orientation (Table 3). In thesecond orientation (score 40.4 compared with 41.5, data not shown), this pair of interactions is inverted. B 13G10/Fe(MoCPP). The uniquecarboxyphenyl substituent is plugged inside the binding site of 13G10, making hydrogen bonds with AsnH33, HisH35 and GlyH100a. One of thephenyl substituents is also inserted into the cavity, while the two others remain solvent exposed. The predicted complex has a similar intermolecularenergy than the 13G10/Fe(ToCPP) complex, with no major bad contacts. C 13G10/aa-Fe(DoCPP). D 13G10/ab-Fe(DoCPP). For the two Fe(DoCPP)derivatives, one of the carboxyphenyl substituents is anchored in the binding site of 13G10, while the other is exposed to the solvent. The buriedcarboxylate makes hydrogen bonds with AsnH33, HisH35, GlyH100 and GlyH100a (Table 3). For aa-Fe(DoCPP), the second carboxylate is interactingwith AlaH97, while in its ab counterpart, it does not make any interaction. As a result, the calculated energy of the aa-Fe(DoCPP)/13G10 complex is2 kJ/mol lower than that of the ab-Fe(DoCPP)/13G10 complex (Table 3). E 14H7/Fe(ToCPP). In the best 14H7/Fe(ToCPP) model, the main hydrogenbonding interactions are formed between the a2 substituent and both SerH97 and TyrH32, and between the a1 substituent and TyrH52 (Table 3). Aweaker interaction occurs between the b substituent and AsnL94 and no major bad contacts are found between any residue and the porphyrincofactor. In other orientations with higher energies, the porphyrin iron is chelated by TyrL32, as observed with Fe(MoCPP). F 14H7/Fe(MoCPP). Theunique carboxylate substituent is interacting with TyrH52, whereas TyrL32 is chelating the metal atom. G 14H7/aa-Fe(DoCPP). H 14H7/ab-Fe(DoCPP).For aa-Fe(DoCPP) and ab-Fe(DoCPP), the interactions of 14H7 with TyrH32 and SerH97, which were present in the 14H7/Fe(ToCPP) complex, aremaintained. Interestingly, in the ab-Fe(DoCPP)/13G10 complex, the interactions with SerH97 and TyrH32 are made with the same carboxylate group,whereas, in aa-Fe(DoCPP)/13G10, one carboxylate substituent is interacting with TyrH32 and the other with SerH97. This leads to the reduction ofbad contacts in the latter case (Table 3).doi:10.1371/journal.pone.0051128.g005

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 10 December 2012 | Volume 7 | Issue 12 | e51128

Ta

ble

3.

Stat

isti

cal

anal

ysis

of

the

low

est

en

erg

yst

ruct

ure

so

bta

ine

din

the

do

ckin

gca

lcu

lati

on

so

fth

ed

iffe

ren

tco

fact

ors

inth

e1

3G

10

and

14

H7

anti

bo

dy

stru

ctu

res.

An

tib

od

yL

iga

nd

Bu

rie

dsu

rfa

cea

rea

(A2

)1S

core

(kJ/

mo

l)2

DG

(kJ/

mo

l)2

Sh

bo

nd

(kJ/

mo

l)2

Hy

dro

ge

nb

on

din

gIn

tera

ctio

ns

(dis

tan

ces

inA

)S

lip

o

(kJ/

mo

l)2

Hy

dro

ph

ob

icco

nta

cts

Sc

las

h

(kJ/

mo

l)2

Cla

she

(dis

tan

ces

inA

)S

inte

rna

l

(kJ/

mo

l)2

Iro

nd

ista

nce

(A)3

13

G1

0Fe

(To

CP

P)

11

9.3

41

.49

24

9.1

33

.38

b-C

OO

2/S

erH

58

OH

(2.9

5)

a1

-C

OO

2/A

snH

33

ND

(3.0

8)

a1

-C

OO

2/ H

isH

35

NE

(2.6

8)

a1

-C

OO

2/ G

lyH

96

NH

(3.0

5)

a1

-C

OO

2/ G

lyH

10

0a

NH

(2.8

2)

27

6.6

5T

yrL3

2,

Asn

L34

,T

rpL9

1,

Ala

L93

,A

snL9

4,

Leu

L96

,A

snH

33

,H

isH

35

,V

alH

50

IleH

51

,T

yrH

52

,A

spH

56

,T

yrH

57

,Se

rH5

8,

Gly

H9

5,

Gly

H9

6,

Ala

H9

7,

Gly

H1

00

a,Ile

H1

00

b,

Me

tH1

00

c

3.0

2C

38

–Se

rH5

8H

B(1

.89

)C

64

–T

rpL9

1H

(2.3

)C

64

–T

rpL9

1C

(3.1

)

4.6

21

.62

aa

–Fe

(Do

CP

P)

13

3.4

42

.01

24

5.2

63

.19

a1

-C

OO

2/ A

snH

33

ND

(2.9

7)

a1

-C

OO

2/ H

isH

35

NE

(2.8

8)

a1

-C

OO

2/ G

lyH

96

N(2

.87

)a

1-

CO

O2

/ Gly

H1

00

aN

(2.9

)a

2-

CO

O2

/ Ala

H9

7N

(3.7

)

24

8.9

Tyr

L32

,T

rpL9

1,

Tyr

H5

2,

Asn

H3

3,

His

H3

5,

Val

H5

0,

IleH

51

,A

spH

56

,T

yrH

57

,Se

rH5

8,

Gly

H9

5,

Gly

H9

6,

Ala

H9

7,

Gly

H1

00

a,Ile

H1

00

b

0.8

7C

70

–G

lyH

10

0aH

A(1

.97

)

2.3

80

.83

ab

–Fe

(Do

CP

P)

12

5.8

74

0.5

12

43

.79

2.6

7a

-C

OO

2/ A

snH

33

ND

(3.0

6)

a-

CO

O2

/ His

H3

5N

E(2

.85

)a

-C

OO

2/ G

lyH

96

N(2

.85

)a

-CO

O2

/ Gly

H1

00

aN

(2.6

)

25

1.0

9T

yrL3

2,

Asn

L34

,T

rpL9

1,

Tyr

H5

2,

Asn

H3

3,

His

H3

5,

Val

H5

0,

IleH

51

,A

spH

56

,T

yrH

57

,Se

rH5

8,

Gly

H9

5,

Gly

H9

6,

Ala

H9

7,

Gly

H1

00

a,Ile

H1

00

b

0.9

2C

58

(H)

–Se

rH5

8H

B2

(2.1

5)

2.3

60

.74

Fe(M

oC

PP

)1

26

.91

42

.12

43

.84

2.6

4C

OO

2/ A

snH

33

ND

(3.0

4)

CO

O2

/ His

H3

5N

E(2

.67

)C

OO

2/ G

lyH

10

0a

N(2

.73

)

25

2.4

6T

yrL3

2,

Asn

L34

,T

rpL9

1,

Tyr

H5

2,

Asn

H3

3,

His

H3

5,

Val

H5

0,

Asp

H5

6,

Tyr

H5

7,

SerH

58

,G

lyH

95

,G

lyH

96

,A

laH

97

,G

lyH

10

0a,

IleH

10

0b

0.2

9–

1.4

61

.75

14

H7

Fe(T

oC

PP

)1

01

.53

31

.54

23

5.6

23

.26

a1

-CO

O2

/ TY

RH

52

OH

(2.6

3)

a2

-CO

O2

/ TY

RH

32

OH

(2.9

6)

a2

-CO

O2

/ SER

H9

7O

H(2

.96

)b

-C

OO

2/ A

SNL9

4N

D(3

.05

)

16

4.5

7T

yrL3

2,

Trp

L91

,A

snL9

4,

Tyr

H3

2,

Asn

H3

3,

His

H3

5,

Tyr

H5

2,

Asn

H5

4,

Asp

H5

6,

Th

rH5

7,

SerH

58

,G

lyH

10

0a,

Leu

H1

00

b

0.2

2–

3.8

7T

yrH

32

OH

(4.5

)

aa

–Fe

(Do

CP

P)

11

6.3

42

9.8

42

32

.08

1.9

8a

1-C

OO

2/ SE

RH

97

OG

(2.5

8)

a2

-CO

O2

/ TY

RH

32

OH

(2.7

6)

17

0.9

6T

yrL3

2,

Trp

L91

,T

yrH

32

,T

yrH

52

,A

spH

56

,G

lyH

10

0a,

Leu

H1

00

b

0.1

2–

2.1

2T

yrH

32

(4.0

1)

ab

–Fe

(Do

CP

P)

10

2.6

62

9.4

32

31

.64

1.9

9a

–C

OO

2/ T

YR

H3

2O

H(2

.73

)a

–C

OO

2/ SE

RH

97

OG

(2.8

0)

16

6.8

5T

yrL3

2,

Trp

L91

,A

snL9

4,

Tyr

H3

2,

Tyr

H5

2,

Asn

H5

4,

Asp

H5

6,

Th

rH5

7,

SerH

58

,G

lyH

10

0a,

Leu

H1

00

b

0.1

9C

70

–SE

RH

58

OH

(2.1

)2

.03

Tyr

H3

2(4

.13

)

Fe(M

oC

PP

)1

09

.55

30

.63

23

2.1

71

.88

CO

O2

/ TY

RH

52

OH

(2.6

)1

74

.54

SerL

30

,T

yrL3

2,

Trp

L91

,T

rpL9

3,

Tyr

H5

2,

SerH

97

,G

lyH

10

0a.

0.1

5–

1.4

Tyr

L32

OH

(2.3

0)

1T

he

bu

rie

dsu

rfac

ear

ea

was

ob

tain

ed

by

sub

trac

tin

gth

em

ole

cula

rsu

rfac

es

(cal

cula

ted

usi

ng

the

UC

SFC

him

era

en

viro

nm

en

t)o

fth

en

on

bo

nd

ed

cofa

cto

ran

do

fth

ean

tib

od

yal

on

e,f

rom

that

of

the

com

ple

xan

db

yd

ivid

ing

the

resu

ltb

y2

.2T

he

Ch

em

Sco

resc

ori

ng

isd

efi

ne

das

:Sc

ore

=2

(DG

+Sc

lash

+Sin

tern

al)

wh

ere

the

tota

lfr

ee

en

erg

ych

ang

eth

ato

ccu

rsu

po

nlig

and

bin

din

gD

G=

25

.48

00

23

.34

00

*Sh

bo

nd

26

.03

00

*Sm

eta

l2

0.1

17

0*S

lip

o+2

.56

00

*H(r

ot)

.S i

nte

rna

lis

the

en

erg

yte

rmfo

rth

ein

tern

alro

tati

on

so

fth

eco

fact

or,

S me

tal

that

for

the

me

tal

inte

ract

ion

san

dH

rot

that

for

the

fro

zen

rota

ble

bo

nd

s.3Fo

r1

3G

10

,d

ista

nce

be

twe

en

the

iro

nat

om

inth

em

od

ele

dco

mp

lex

and

Mg

2+

inth

ecr

ysta

lst

ruct

ure

.Fo

r1

4H

7,

dis

tan

ceto

iro

n.

do

i:10

.13

71

/jo

urn

al.p

on

e.0

05

11

28

.t0

03

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 11 December 2012 | Volume 7 | Issue 12 | e51128

deprotonating the substrate and chelating the metal [50]. For the

peroxidase reaction, AspH96 is too near to the center of the

porphyrin ring to act as a distal ligand but it is in an appropriate

position to act as a proximal ligand for the iron atom of Fe(III)-

mesoporphyrin IX. Hydrogen peroxide is expected to approach

the non-obstructed side of the porphyrin ring, opposite to AspH96,

which is surrounded only by hydrophobic residues such as TyrL49

and TyrL91. The active site of 7G12 does not reveal the presence

of distal ligands susceptible to enhance the peroxidase activity of

Fe(III)mesoporphyrin IX.

Directed and random mutagenesis were used to generate

libraries of 7G12, and antibodies with increased peroxidase

activity were selected by phage display using an activity-based

strategy [53]. A mutant, in which TyrL49 was replaced by Trp,

had a 10-fold increase in kcat/Km. In addition, two mutants bore

mutations thought to affect the packing between the porphyrin

ring and TyrL49 or TyrL91. Because TyrL49 and TrpL91 are

involved in p-stacking interactions with the porphyrin ring in the

7G12/N-methylmesoporphyrin IX complex (Fig. 8A), it was

proposed that the mutations could help to stabilize the radical

cation on the porphyrin ring and yield to higher peroxidase

activity [53].

In 13G10 and 7G12, a bulky residue preceding residue H101

(Met) is positioned at the bottom of the active site cavity, which

leads to shallow binding sites (Fig. 8). This positions the porphyrin

haptens at the surface of the combining site, contacting almost

exclusively residues of the CDRs. The larger elbow angle in l-light

chain antibody 13G10 coupled with a longer CDRH3 (H95–

H101) results in a shallower binding site in 13G10 compared with

7G12 (Fig. 8A and B). In contrast to 7G12, no residue is correctly

positioned to bind the iron in the 13G10-Fe(ToCPP) complex.

Discussion

Antibodies elicited against carefully designed transition state

analogues have been reported to catalyze a wide range of chemical

reactions [54,55,56,57,58,59]. However, the crystal structures of

these antibody catalysts have revealed that most antibodies

generally act by simple transition state stabilization [58,60,61]

and only a few utilize covalent chemistry [62,63,64,65]. The

incorporation of cofactors has been proposed as a possible strategy

to expand the catalytic scope of antibodies, in particular to lead to

redox catalysis [56]. Actually, the combination of the intrinsic

reactivity of a cofactor with the tailored binding specificity of an

antibody has been underutilized. However, cofactor and antibody

effectively complement each other: like in enzymatic catalysis, the

antibody enhances the catalytic efficiency of the cofactor and

ensures reaction specificity, stereospecificity and substrate speci-

ficity. Several antibodies raised against a porphyrin hapten and

possessing catalytic activity have been reported

[12,48,49,66,67,68,69,70,71,72]. Moreover, catalytic antibody-

cofactor complexes have been structurally characterized for a

periodate-dependent oxygenation catalyst [73,74], and a pyridox-

al-59-phosphate-dependent antibody that catalyzes the transami-

nation of D-amino acids [75].

Antibodies 13G10 and 14H7, induced against Fe(ToCPP), were

shown to display peroxidase activity in the presence of iron-

porphyrin cofactors. Biochemical studies had already given

insights into the interactions between the antibodies and the

cofactor. First, UV-visible studies have shown that the binding of

the porphyrin into the antibodies is not accompanied by a change

of the high spin state of the iron(III) and that the porphyrin binds

in a hydrophobic pocket [13]. Moreover, the antibodies possess

similar affinities for the metallated or non-metallated cofactor

(Kd = 3–5 nM) [14]. Altogether, these results indicated that no

amino acid binding the iron atom had been induced in the

antibody combining sites by the Fe(ToCPP) hapten. The

determination of the apparent dissociation constants for the

variously substituted porphyrins by competitive ELISA indicated

that the antibodies do not bind tetraphenylporphyrin and allowed

a model to be proposed, where two thirds of the porphyrin

macrocycle could be inserted in the binding pocket, with two

carboxylates in a,b positions being more specifically bound to the

protein [14]. Tetraaryl porphyrins bearing only one meso-ortho-

carboxyphenyl substituent could still bind 13G10 and 14H7,

although with a 50-fold reduction in affinity [14]. Finally,

absorption spectroscopy studies have shown that, whereas the

iron(III) of Fe(ToCPP) is able to bind two imidazole ligands, the

Fe(ToCPP)-13G10 complex can fix only one, which inhibits its

peroxidase activity [16].

All catalytic antibodies, whose crystal structure had been solved

until now, belonged to the 95% mouse antibodies that possess a k-

light chain [76], except one [77]. Many of these catalytic

antibodies share a deep combining site formed not only by

residues of the CDRs but also by residues of the framework

(Figure 8C) [58,61]. This observed structural convergence was due

in part to the use of similar hydrophobic haptens to induce the

antibodies. 13G10 and 14H7, elicited against a very different iron-

porphyrin hapten, belong to the IgG1, l class. As observed for

other catalytic antibodies [78], the mature genes of 13G10 and

14H7 display a high conservation degree with their germline

counterparts.

Among the few examples of crystal structures of murine Fabs

with l-type light-chain that have been solved, ten of them are

those of Fabs complexed with small haptens

[22,79,80,81,82,83,84,85] (Table 2) and several of them with a

peptide or protein antigen [86,87,88,89,90,91,92,93]. These

crystal structures have shown that the combining site of l-light

chain antibodies is formed by different amino acid residues than

that of k-light chain antibodies (Compare Table 2 with Table 2 of

reference [61]). In particular, the X-ray structures of antibodies of

different isotypes (l– and k– type) that bind to the same hapten

molecule have revealed drastically different binding modes of the

ligand [83,84,94]. It was therefore anticipated that catalytic

antibodies 13G10 and 14H7 will possess a combining site different

from that of previously crystallographically characterized k–light

chain catalytic antibodies and it was interesting to understand how

different they were.

The comparison of the cavity shapes of non hydrolytic catalytic

antibodies led to separate them into two categories, depending on

the nature of the residue preceding H101 [61]. Antibodies

possessing a small residue (ie. Gly, Ser) had a common deep

combining site, whereas those with a bulky residue (Phe, Met)

displayed shallow cavities with very different shapes. Antibodies

13G10 and 14H7 differ at this position, possessing Met and Leu,

respectively, and their structures reveal that their combining site

cavities is not very deep. In addition, their combining sites lie more

on the protein surface, compared with the other catalytic

antibodies possessing a bulky residue preceding H101

(Figure 8C). Because CDRH3 that is composed of several glycines

and residues with small side chains is predicted to be flexible, the

combining sites of the antibodies 13G10 and 14H7 do not share a

high similarity in shape although the two antibodies share a high

sequence similarity. Indeed, the deeper cavity observed in the

structure of 13G10 compared with 14H7 comes mainly from the

different conformations of CDRH3.

The structures of 13G10 and 14H7 indicate that the amino

acids prone to bind the ligand are the same as in the other l-light

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 12 December 2012 | Volume 7 | Issue 12 | e51128

Figure 6. Molecular surfaces of the models of the complexes of 13G10 and 14H7 with the different porphyrins. Residues involved invan der Waals and hydrophobic contacts are shown. A 13G10/Fe(ToCPP). B 13G10/Fe(MoCPP). C 13G10/aa-Fe(DoCPP). D 13G10/ab-Fe(DoCPP). E14H7/Fe(ToCPP). F 14H7/Fe(MoCPP). G 14H7/aa-Fe(DoCPP). H 14H7/ab-Fe(DoCPP).doi:10.1371/journal.pone.0051128.g006

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 13 December 2012 | Volume 7 | Issue 12 | e51128

chain antibodies (Table 2), with only a few residues that belong to

the L chain taking part in the combining site, and CDRH2 making

an important contribution. About half of the hapten appears to be

buried in the combining sites of 13G10 and 14H7, with CDRH3

and residues TyrL32, TyrH52 and TrpL91 stacking against the

porphyrin ring (Figure 6). Although this accounts for the nM

affinity of the hapten for the antibodies, this is lower than the two

thirds predicted from the biochemical data [14], presumably

because induced fit is expected to occur in the complexes, which is

not taken into account in the calculated interactions. Molecular

docking predicts important differences for the recognition of

Fe(ToCPP) by the two antibodies. The shallower and wider

binding site of 14H7, as observed in the crystal structure, does not

allow the hapten to bind as deeply as it does in 13G10. Similar

results were obtained for the docking of the other cofactors

(Figure 7A and 7B).

Like in peroxidases, the heme environment of 13G10 and 14H7

consists of numerous hydrophobic residues. In heme peroxidases,

the iron atom is bound to the apoprotein by the proximal histidine.

This axial ligand was shown to be important for modulating the

redox potential of the iron and thus, the catalytic activity of the

enzyme [95]. However, our docking models indicate, in agreement

with the biochemical data, that no amino acid residue that binds

the iron has been induced in the combining sites of 13G10 and

14H7. The key step of the mechanism of peroxidases is usually the

cleavage of the O–O bond of H2O2 or ROOH, with the release of

a water molecule and formation of the highly reactive iron (V)-oxo

intermediate, assisted by the distal histidine and an arginine.

However, this function is fulfilled by a catalytic glutamate base

positioned 4.9 A apart from the iron for a direct attack in

Caldariomyces fumago chloroperoxidase [96]. Neither induction by

the Fe(ToCPP) hapten itself, nor the use of differently substituted

porphyrin cofactors, led to the positioning of any histidine or

arginine residue in 13G10 or 14H7 that could participate in the

heterolytic cleavage of the O-O bond of peroxide, like in

peroxidases. However, previous biochemical results suggested that

a carboxylic acid side chain of antibody 13G10 could participate

in catalysis by protonating one of the oxygen atom of H2O because

it was shown that kcat increases sharply below pH 5 for the 13G10-

Fe(ToCCP) complex [15]. However, no COOH side-chain of the

13G10 antibody is correctly localized to act as a general acid-base

catalyst. Indeed, the only carboxylic acid side-chain in the

combining site of 13G10 that could play a role in catalysis

belongs to AspH56 (Table 3, Fig. 3C, 4C, 5, 6 and 8A). However,

its nearest oxygen atom is positioned 10.6 A away from the iron, a

distance that necessitates a mediating water molecule for AspH56

to act as a general acid catalyst in the peroxidase reaction. Acid

catalysis by the ortho carboxylate group of iron (II) [meso-(ortho-

carboxyphenyltriphenyl-porphyrin has previously been proposed

in aldoxime dehydration via the protonation of the substrate

hydroxy group [97]. Thus, one of the two buried carboxylate

groups of the hapten could fulfill the same function in the 13G10-

Fe(ToCCP) complex. Our models of the porphyrin-13G10

complexes reveal that, in addition to them, the polar residues

best positioned to play a role in catalysis are AsnH33 and TyrH52,

also located on the buried side of the porphyrin (Figure 5 A–D).

H2O2 could bind on the same sheltered face of the porphyrin ring,

Figure 7. Comparison of the binding of the different porphy-rins to 13G10 and 14H7 and models of imidazole binding tothe 13G10/porphyrin complexes. Stereoviews of the superpositionof the different porphyrins in the combining sites of 13G10 (A) and14H7 (B). Fe(ToCPP), aa-Fe(DoCPP), ab-Fe(DoCPP) and Fe(MoCPP) arecolored green, blue, yellow and red, respectively. For 13G10, theporphyrins of Fe(MoCPP), aa-Fe(DoCPP) and ab-Fe(DoCPP) superim-pose onto that of Fe(ToCPP)(without taking into account thecarboxylate substituents) with rmsds of 0.5, 3.0 and 1.7 A2, respectively.For 14H7, the porphyrins of Fe(MoCPP), aa-Fe(DoCPP) and ab-Fe(DoCPP) superimpose onto that of Fe(ToCPP) with rmsds of 9.1, 3.1and 0.9 A2, respectively. C Model of imidazole bound to 13G10/Fe(ToCPP). The binding of the imidazole molecule is favored by an H-bonding interaction with the main chain of AlaL93. D Model ofimidazole bound to 13G10/aa-Fe(DoCPP). Despite some clashes,imidazole binding is stabilized by H-bonds with AsnH33 and one ofthe buried carboxylates. Stabilization is also achieved by hydrophobic

contacts as imidazole binds in the cavity left between the cofactor andthe protein. E Model of imidazole bound to 13G10/ab-Fe(DoCPP).Imidazole is located on the buried side of the porphyrin but does notmake any interaction with the protein, the b-carboxylate being orientedtoward the solvent.doi:10.1371/journal.pone.0051128.g007

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 14 December 2012 | Volume 7 | Issue 12 | e51128

in a hydrophobic pocket similar to that present in horseradish

peroxidase [98]. This location of the binding site of hydroperoxide

would explain the remarkable thermostability of the cofactor-

antibody complex and the multiple turnovers of the reaction. In

cytochrome c oxidase, the radical cation in the iron(IV)oxopor-

phyrin intermediate was delocalized onto the indole ring of

TrpL91 [8]. Similarly, it was proposed that TyrL49 and TyrL91

that stack against the porphyrin ring of the hapten could fulfill a

similar function in catalytic antibody 7G12 and enhance the

peroxidase activity of the 7G12/Fe(III)mesoporphyrin complex

[53]. In the models of 13G10 and 14H7 with the different

porphyrins, TrpL91, TyrL32 and TyrH52, which are predicted to

stack against the porphyrin ring (Fig. 6), could act in the same way.

Imidazole was used as an iron ligand that might mimic the

proximal histidine in the 13G10/porphyrin complexes. Molecular

docking indicates that imidazole preferentially binds on opposite

faces of the porphyrin ring in the 13G10/Fe(ToCPP) and 13G10/

Fe(DoCPP) complexes (Figure 7C to E). Imidazole is predicted to

bind to the solvent-exposed face of Fe(ToCPP), which might

hinder binding of the hydroperoxide substrate to the sheltered face

of the cofactor and lead to the inhibition of the peroxidase activity.

In contrast, imidazole is predicted to bind in the cavity formed

between 13G10 and aa- or ab-Fe(DoCPP) (Figure 7D and E),

which likely represents a higher affinity site and would explain that

the affinity for imidazole of the a,a- and a,b-1,2-Fe(DoCPP)

complexes was 2–3 fold lower than that of 13G10/Fe(ToCPP). In

this manner, imidazole would be located appropriately to act as

the proximal histidine. In this case, hydroperoxide would bind on

the solvent-exposed face of the porphyrin and TyrL32 could assist

in the cleavage of the O-O bond of hydroperoxide by functioning

as an acid catalyst. In the case of ab-Fe(DoCPP), the b carboxylate

of could also play this role. This would account for the 8–9–fold

enhancement of the catalytic efficiency of the peroxidase reaction

in the presence of 50 mM imidazole, when 13G10 is complexed

with the di-substituted cofactors compared with Fe(ToCPP).

In the future, better catalysts could be obtained by directed

mutagenesis of 13G10 and 14H7. Replacing AsnH33 by a

histidine in 13G10 could lead to the binding of its imidazole group

to the iron atom of Fe(ToCPP), which could enhance the

peroxidase activity of the 13G10/porphyrin complex. Indeed,

antibodies induced against microperoxidase 8 that were shown to

possess an axial histidine coordinating the iron atom had a better

peroxidase activity than 13G10 and 14H7 [99,100]. Mutating

TyrL32 to histidine or arginine to generate a better acid catalyst or

an amino acid that enhances the polarization of the O-O bond

could also increase the catalytic activity of the 13G10/

Fe(DoCPP)/imidazole complexes. Alternatively, antibodies with

enhanced peroxidase activity could be obtained by phage display

using an activity-based strategy for selecting oxidative catalysts, as

described previously [53].

ConclusionUnderstanding the structure-function relationship of catalytic

antibodies with a l-light chain is important to shed light on the

diversity of catalytic antibodies, which may help to broaden the

scope of these catalysts. The anti-porphyrin antibodies 13G10 and

14H7, with a l-light chain, were shown to possess shallow hapten

binding pockets compared with the other structurally character-

ized catalytic antibodies. The structural complementarity of the

Fe(ToCPP) cofactor to the hydrophobic binding pocket of

antibodies 13G10 and 14H7 leads to a remarkable thermostability

of the cofactor-antibody complexes and allows multiple turnovers

of the peroxidase reaction. Molecular modeling indicates that the

recognition of various porphyrins with different carboxyphenyl

substituents is achieved mainly by stacking interactions but also by

crucial hydrogen bonds with two or three carboxylate groups. Our

models explain why one carboxyphenyl substituent is sufficient for

a good affinity of the porphyrin cofactor for 13G10 and 14H7.

CDRH1 and CDRH3 appear to play key roles for binding

Fe(ToCPP) and no proximal ligand of the iron was induced in

13G10 and 14H7. The increase of the peroxidase activity of the

cofactor, when bound to the antibodies, could be explained by a

loss of entropy due to accessibility of H2O2 to only one of the two

faces of the porphyrin ring, and possibly by the stacking of several

aromatic groups onto the porphyrin ring that would stabilize the

radical cation in the iron(IV)oxoporphyrin intermediate in the

peroxidase reaction. Moreover, one of the buried carboxylic group

of the porphyrin substituents may also participate as a general acid

catalyst, with the implication that H2O2 would bind on the

sheltered face of the porphyrin ring. The carboxylic group would

facilitate the heterolytic cleavage of the O-O bond, with the release

of a water molecule and formation of the highly reactive iron (V)-

oxo intermediate. Interestingly, an imidazole ligand that mimics

the proximal histidine in peroxidases can be modeled in the

complexes of 13G10 with aa-Fe(DoCPP) and ab-Fe(DoCPP) on

the sheltered face of the porphyrin ring, with TyrL32 correctly

positioned to act as an acid catalyst, in agreement with these

complexes showing higher catalytic efficiency compared with

Fe(ToCPP), in the presence of imidazole. The similar binding

energies of 13G10 and 14H7 for the alternative cofactors and the

original hapten open novel possibilities for developing other

porphyrinic-based cofactors for these antibodies. In future work, in

order to obtain catalytic antibodies that mimic cytochromes P450,

one should design a substituted porphyrin hapten that generates

binding sites for the substrate in addition to that of the

metalloporphyrin in the antibodies.

Table 4. Statistical analysis of the lowest energy structures for the docking of imidazole in the 13G10/Fe(ToCPP) and 13G10/Fe(DoCPP) models.

Antibody Ligand ScoreDGbinding

(kJ/mol) Shbond

Hydrogen bondingInteractions(distances in A) Smetal Slipo Sclash

Clashes(distancesin A)

13G10 Fe(ToCPP) 22.61 223.08 0.98 AlaL93-O (2.64) 0.9 76.42 0.47 –

aa–Fe(DoCPP) 21.54 222.87 0.63 COO2 (2.65) 0.81 88.96 1.33 AsnH33 (2.45)

AsnH33 ND (2.96) Fe (2.62)

ab–Fe(DoCPP) 19.88 219.93 0.0 – 0.9 77.21 0.05 –

doi:10.1371/journal.pone.0051128.t004

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 15 December 2012 | Volume 7 | Issue 12 | e51128

Figure 8. Comparison of Fab 13G10 with other catalytic antibodies. A Comparison of the combining sites of the two porphyrin-dependentperoxidase antibodies 13G10 and 7G12. The a-carbons of the framework of the heavy chain variable domain of 7G12 (in orange) have beensuperimposed on those of 13G10 (in yellow), which results in a large variation in the positions of the light chain variable domains (VL 7G12 inmagenta and VL 13G10 in pink). The rms deviation of 0.964 A reflects the different packing of the VH and VL domains in l-light chain antibodies(13G10) and k-light chain antibodies (7G12). The hapten of 13G10 is represented in blue and that of 7G12 in red. B Comparison of the cavitydeepness of antibodies 13G10 (in blue) and 7G12 (in orange). A section of the molecular surface of the combining site of antibody 13G10 is colored inyellow for the heavy chain and pink for the light chain. C Comparison of the cavity shapes of the nonhydrolytic catalytic antibodies with a bulkyresidue at position preceding H101. 15A9 (yellow), PDB code 2BMK; 28B4 (red), PDB code 1KEL; 1E9 (green), PDB code 1C1E; 34D4 (cyan), PDB code1Y18; 1D4 (magenta), PDB code 1JGU.doi:10.1371/journal.pone.0051128.g008

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 16 December 2012 | Volume 7 | Issue 12 | e51128

Data deposition. The atomic coordinates and structure

factors of Fab13G10 and Fab14H7 have been deposited at the

Protein Data Bank (PDB codes 4amk and r4amksf, and 4at6 and

r4at6sf, respectively).

Supporting Information

Figure S1 Detection of twinning and determination ofthe twin fraction in the 14H7 crystals. A Estimation of the

twin fraction a by Britton plot analysis. The percentage of negative

intensities after detwinning is plotted as a function of the assumed

value of a. The estimated value of a is extrapolated from the linear fit

(green line). B Estimation of the twin fraction a using the H-plot. The

cumulative fractional intensity difference of acentric twin-related

intensities H {H = |I(h 1) 2 I(h 2)|/[I(h 1) + I(h 2)]} is plotted against

H. The initial slope (green line) of the distribution is a measure of a.

(DOCX)

Acknowledgments

We thank Diep Le for the N-terminal sequencing, Benoıt Gigant for

reading the manuscript, and Remy Ricoux for initiating the collaboration

between B. G.-P. and J.-D.M.

Author Contributions

Conceived and designed the experiments: JDM JPM BGP. Performed the

experiments: VMR AB MAS BGP. Analyzed the data: VMR JDM JPM

BGP. Wrote the paper: JDM MAS BGP.

References

1. Smith LJ, Kahraman A, Thornton JM (2010) Heme proteins–diversity in

structural characteristics, function, and folding. Proteins 78: 2349–2368.

2. Mansuy D, Battioni P (1994) Cytochrome P450 model systems. In: Dekker M,

editor. MetaIloporphyrins in Catalytic Oxidations. New York: R.A. Sheldon.

99–132.

3. Mahy J-P, Desfosses B, de Lauzon S, Quilez R, Desfosses B, et al. (1998)

Hemoabzymes: Different strategies for obtaining artificial hemoproteins basedon antibodies. Applied Biochem Biotech 75: 103–112.

4. Ricoux R, Sauriat-Dorizon H, Girgenti E, Blanchard D, Mahy JP (2002)Hemoabzymes: towards new biocatalysts for selective oxidations. J Immunol

Methods 269: 39–57.

5. Mahy JP, Raffy Q, Allard M, Ricoux R (2009) Various strategies for obtaining

artificial hemoproteins: from ‘‘hemoabzymes’’ to ‘‘hemozymes’’. Biochimie 91:1321–1323.

6. Robertson DE, Farid RS, Moser CC, Urbauer JL, Mulholland SE, et al. (1994)Design and synthesis of multi-haem proteins. Nature 368: 425–432.

7. Shinde S, Cordova JM, Woodrum BW, Ghirlanda G (2012) Modulation of

function in a minimalist heme-binding membrane protein. J Biol Inorg Chem17: 557–564.

8. Millett F, Miller MA, Geren L, Durham B (1995) Electron transfer betweencytochrome c and cytochrome c peroxidase. J Bioenerg Biomembr 27: 341–351.

9. Ortiz de Montellano PR (1992) Catalytic sites of hemoprotein peroxidases. AnnuRev Pharmacol Toxicol 32: 89–107.

10. Marnett LJ, Kennedy TA (1995). Cytochrome P450: Structure, mechanism andBiochemistry: Ortiz de Montellano, P.R., 2nd Ed., Plenum Press, New York,

London. 49–80.

11. Ricoux R, Dubuc R, Dupont C, Marechal JD, Martin A, et al. (2008)

Hemozymes peroxidase activity of artificial hemoproteins constructed from theStreptomyces lividans xylanase A and iron(III)-carboxy-substituted porphyrins.

Bioconjug Chem 19: 899–910.

12. Nimri S, Keinan E (1999) Antibody-metalloporphyrin catalytic assembly mimics

natural oxydation enzymes. J Am Chem Soc 121: 8978–8982.

13. Quilez R, de Lauzon S, Desfosses B, Mansuy D, Mahy J-P (1996) Artificial

peroxidase-like hemoproteins based on antibodies constructed from a specifically

designed ortho-carboxy substituted tetraarylporphyrin hapten and exhibiting ahigh affinity for iron-porphyrins. FEBS Lett 395: 73–76.

14. de Lauzon S, Quilez R, Lion L, Desfosses B, Desfosses B, et al. (1998) Active sitetopology of artificial peroxidase-like hemoproteins based on antibodies

constructed from a specifically designed ortho-carboxy-substituted tetraarylpor-phyrin. Eur J Biochem 257: 121–130.

15. de Lauzon S, Desfosses B, Mansuy D, Mahy J-P (1999) Studies of the reactivityof artificial peroxidase-like hemoproteins based on antibodies elicited against a

specifically designed ortho-carboxy substituted tetraarylporphyrin. FEBS Lett

443: 229–234.

16. de Lauzon S, Mansuy D, Mahy JP (2002) Coordination chemistry of iron(III)-

porphyrin-antibody complexes. Eur J Biochem 269: 470–480.

17. Jones ST, Bendig MM (1991) Rapid PCR-cloninng of full-length mouse

immunoglobulin variable regions. Biotechnology 9: 88–89 & 579.

18. Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, et al. (1988) Primer-

directed enzymatic amplification of DNA with a thermostable DNA ploymerase.Science 239: 487–491.

19. Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci 74: 5463–5467.

20. Otwinovsky Z, Minor W (1997) Processing of X-ray diffraction data collected inoscillation mode. Methods Enzymol 276: 307–325.

21. Navaza J (1994) AMoRe: an automated package for molecular replacement.Acta Crystallogr A 50: 157–163.

22. Cygler M, Rose DR, Bundle DR (1991) Recognition of a cell-surfaceoligosaccharide of pathogenic Salmonella by an antibody Fab. Science 253:

442–445.

23. Jones TA, Zhou J-Y, Cowan SW, Kjeldgaard M (1991) Improved methods for

building protein models in electron density maps and the location of errors inthese models. Acta Crystallogr A 47: 110–119.

24. Adams PD, Pannu NS, Read RJ, Brunger AT (1997) Cross-validated maximum

likelihood enhances crystallographic simulated annealing refinement. Proc Natl

Acad Sci U S A 94: 5018–5023.

25. Adams PD, Afonine PV, Bunkoczi G, Chen VB, Davis IW, et al. (2010)

PHENIX: a comprehensive Python-based system for macromolecular structuresolution. Acta Crystallogr D Biol Crystallogr 66: 213–221.

26. Becke AD (1993) Densityfunctional thermochemistry. III. The role of exactexchange. J Chem Phys 98: 5648–5652.

27. Lee C, Yang W, Parr RG (1988) Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys Rev B: Condens

Matter 37: 785–789.

28. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA, et al. (2009).

Wallingford CT: Gaussian Inc.

29. Hay PJ, Wadt WR (1985) Ab initio effective core potentials for molecular

calculations. Potentials for K to Au including the outermost core orbitals. J ChemPhys 82: 299–310.

30. Hehre WJ, Ditchfield R, Pople JA (1972) Self–Consistent Molecular Orbital

Methods. XII. Further Extensions of Gaussian–Type Basis Sets for Use inMolecular Orbital Studies of Organic Molecules. J Chem Phys 56: 2257–2261.

31. Hariharan PC, Pople JA (1973) The influence of polarization functions onmolecular orbital hydrogenation energies. Theor Chim Acta 28: 213–222.

32. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, et al.(2004) UCSF Chimera–a visualization system for exploratory research and

analysis. J Comput Chem 25: 1605–1612.

33. Munoz Robles V, Ortega-Carrasco E, Gonzalez Fuentes E, Lledos A, Marechal

JD (2011) What can molecular modelling bring to the design of artificialinorganic cofactors? Faraday Discuss 148: 137–159; discussion 207–128.

34. Seebeck B, Reulecke I, Kamper A, Rarey M (2008) Modeling of metalinteraction geometries for protein-ligand docking. Proteins 71: 1237–1254.

35. Eldridge MD, Murray CW, Auton TR, Paolini GV, Mee RP (1997) Empiricalscoring functions: I. The development of a fast empirical scoring function to

estimate the binding affinity of ligands in receptor complexes. J Comput Aided

Mol Des 11: 425–445.

36. Dunbrack RL, Jr. (2002) Rotamer libraries in the 21st century. Curr Opin Struct

Biol 12: 431–440.

37. North B, Lehmann A, Dunbrack RL, Jr. (2011) A new clustering of antibody

CDR loop conformations. J Mol Biol 406: 228–256.

38. Ganesh K, Neuberger MS (2011) The relationship between hypothesis and

experiment in unveiling the mechanisms of antibody gene diversification.FASEB J 25: 1123–1132.

39. Strausberg RL, Feingold EA, Grouse LH, Derge JG, Klausner RD, et al. (2002)Generation and initial analysis of more than 15,000 full-length human and

mouse cDNA sequences. Proc Natl Acad Sci U S A 99: 16899–16903.

40. Kabat EA, Wu TT, Perry HM, Gottesman KS, Foeller C (1991) Sequences of

Proteins of Immunological interest. Bethesda, MD, ed. 5: U. S. Public Health Service,National Institutes of Health.

41. Chang S, Mohan C (2005) Identification of novel VH1/J558 immunoglobulingermline genes of C57BL/6 (Igh b) allotype. Mol Immunol 42: 1293–1301.

42. Larsen NA, Heine A, de Prada P, Redwan el R, Yeates TO, et al. (2002)Structure determination of a cocaine hydrolytic antibody from a pseudomer-

ohedrally twinned crystal. Acta Cryst D58: 2055–2059.

43. Golinelli-Pimpaneau B (2005) Structure of a pseudomerohedrally twinned

monoclinic crystal form of a pyridoxal phosphate-dependent catalytic antibody.

Acta Crystallogr D Biol Crystallogr 61: 472–476.

44. Hamdane D, Lechauve C, Marden MC, Golinelli-Pimpaneau B (2009) Pseudo-

merohedral twinning in monoclinic crystals of wild-type human brainneuroglobin. Acta Crystallogr D Biol Crystallogr 65: 388–392.

45. Padilla JE, Yeates TO (2003) A statistic for local intensity differences: robustnessto anisotropy and pseudo-centering and utiliy for detecting twinning. Acta

Crystallogr D Biol Crystallogr 59: 1124–1130.

46. Schaefer WH, Harris TM, Guengerich FP (1985) Characterization of the

enzymatic and nonenzymatic peroxidative degradation of iron porphyrins andcytochrome P-450 heme. Biochemistry 24: 3254–3263.

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 17 December 2012 | Volume 7 | Issue 12 | e51128

47. Britton D (1972) Estimation of twinning parameter for twins with exactly

superimposed reciprocal lattices. Acta Cryst A28: 296–297.

48. Cochran AG, Schultz PG (1990) Antibody-catalyzed porphyrin metallation.

Science 249: 781–783.

49. Cochran AG, Schultz PG (1990) Peroxidase activity of an antibody-hemecomplex. J Am Chem Soc 112: 9414–9415.

50. Romesberg FE, Santarsiero BD, Spiller B, Yin J, Barnes D, et al. (1998)Structural and kinetic evidence for strain in biological catalysis. Biochemistry 37:

14404–14409.

51. Yin J, Andryski SE, Beuscher AEt, Stevens RC, Schultz PG (2003) Structuralevidence for substrate strain in antibody catalysis. Proc Natl Acad Sci U S A 100:

856–861.

52. Yin J, Beuscher AEt, Andryski SE, Stevens RC, Schultz PG (2003) Structural

plasticity and the evolution of antibody affinity and specificity. J Mol Biol 330:

651–656.

53. Yin J, Mills JH, Schultz PG (2004) A catalysis-based selection for peroxidase

antibodies with increased activity. J Am Chem Soc 126: 3006–3007.

54. Golinelli-Pimpaneau B, Goncalves O, Dintinger T, Blanchard D, Knossow M,

et al. (2000) Structural evidence for a programmed general base in the active site

of a catalytic antibody. Proc Natl Acad Sci U S A 97: 9892–9895.

55. Stevenson JD, Thomas NR (2000) Catalytic antibodies and other biomimetic

catalysts. Nat Prod Rep 17: 535–577.

56. Hilvert D (2000) Critical analysis of antibody catalysis. Annu Rev Biochem 69:

751–793.

57. Wentworth P, Jr., Janda KD (2001) Catalytic antibodies: structure and function.

Cell Biochem Biophys 35: 63–87.

58. Golinelli-Pimpaneau B (2002) Structural diversity in antibody catalysts.J Immunol Methods 269: 157–171.

59. Xu Y, Yamamoto N, Janda KD (2004) Catalytic antibodies: hapten designstrategies and screening methods. Bioorg Med Chem 12: 5247–5268.

60. Charbonnier JB, Gigant B, Golinelli-Pimpaneau B, Knossow M (1997)

Similarities of hydrolytic antibodies revealed by their X-ray structures: a review.Biochimie 79: 653–660.

61. Golinelli-Pimpaneau B (2000) Novel reactions catalysed by antibodies. Curr OpStruc Biol 10: 697–708.

62. Wagner J, Lerner RA, Barbas III CF (1995) Efficient aldolase catalytic

antibodies that use the enamine mechanism of natural enzymes. Science 270:1797–1800.

63. Thayer MM, Olender EH, Arvai AS, Koike CK, Canestrelli IL, et al. (1999)Structural basis for amide hydrolysis catalyzed by the 43C9 antibody. J Mol Biol

291: 329–345.

64. Tanaka F, Barbas CF, 3rd (2002) Reactive immunization: a unique approach tocatalytic antibodies. J Immunol Methods 269: 67–79.

65. Goswami RK, Huang ZZ, Forsyth JS, Felding-Habermann B, Sinha SC (2009)Multiple catalytic aldolase antibodies suitable for chemical programming. Bioorg

Med Chem Lett 19: 3821–3824.

66. Savitsky AP, Nelen MI, Yatsmirsky AK, Demcheva MV, Ponomarev GV, et al.(1994) Kinetics of oxidation of o-dianisidine by hydrogen peroxide in the

presence of antibody complexes of iron(III) coproporphyrin. Appl BiochemBiotechnol 47: 317–327.

67. Takagi M, Kohda K, Hamuro T, Harada A, Yamaguchi H, et al. (1995)Thermostable peroxidase activity with a recombinant antibody L chain-

porphyrin Fe(III) complex. FEBS Lett 375: 273–276.

68. Feng Y, Liu Z, Gao G, Gao SJ, Liu XY, et al. (1995) Study of the abzyme withperoxidase catalytic activity. Ann N Y Acad Sci 750: 271–276.

69. Kawamura-Konishi Y, Hosomi N, Neya S, Sugano S, Funasaki N, et al. (1996)Kinetic characterization of antibody-catalyzed insertion of a metal ion into

porphyrin. J Biochem 119: 857–862.

70. Harada A, Fukushima H, Shiotsuki K, Yamaguchi H, Oka F, et al. (1997)Peroxidation of Pyrogallol by Antibody-Metalloporphyrin Complexes. Inorg

Chem 36: 6099–6102.

71. Kohda K, Kakehi M, Ohtsuji Y, Tagaki M, Imanaka T (1997) Studies of high

thermostability and peroxidase activity of recombinant antibody L chain-

porphyrin Fe(III) complex. FEBS Lett 407: 280–284.

72. Hosomi N, Kawamura-Konishi Y, Kawano R, Fujii I, Suzuki H (2005) Site-

directed mutagenesis study of the antibody 2D7 which catalyzes a reaction forinsertion of Cu2+ into mesoporphyrin. J Biosci Bioeng 99: 222–229.

73. Hsieh-Wilson LC, Schultz PG, Stevens RC (1996) Insights into antibody

catalysis: Structure of an oxygenation catalyst at 1.9 A resolution. Proc NatlAcad Sci USA 93: 5363–5367.

74. Yin J, Mundorff EC, Yang PL, Wendt KU, Hanway D, et al. (2001) Acomparative analysis of the immunological evolution of antibody 28B4.

Biochemistry 40: 10764–10773.

75. Golinelli-Pimpaneau B, Luthi C, Christen P (2006) Structural basis for D-aminoacid transamination by the pyridoxal 59-phosphate-dependent catalytic antibody

15A9. J Biol Chem 281: 23969–23977.

76. Eisen HN, Reilly EB (1985) Lambda chains and genes in inbred mice. Annu Rev

Immunol 3: 337–365.77. Larsen NA, de Prada P, Deng SX, Mittal A, Braskett M, et al. (2004)

Crystallographic and biochemical analysis of cocaine-degrading antibody

15A10. Biochemistry 43: 8067–8076.78. Le Minoux D, Mahendra A, Kaveri S, Limnios N, Friboulet A, et al. (2012) A

novel molecular analysis of genes encoding catalytic antibodies. Mol Immunol50: 160–168.

79. Love RA, Villafranca JE, Aust RM, Nakamura KK, Jue RA, et al. (1993) How

the anti-(metal chelate) antibody CHA255 is specific for the metal ion of itsantigen: X-ray structures for two Fab’/hapten complexes with different metals in

the chelate. Biochemistry 32: 10950–10959.80. Yuhasz SC, Parry C, Strand M, Amzel LM (1995) Structural analysis of affinity

maturation: The three-dimensional structures of complexes of an anti-nitrophenol antibody. Molec Immunol 32: 1143–1155.

81. Mizutani R, Miura K, Nakayama T, Shimada I, Arata Y, et al. (1995) Three-

dimensional structures of the Fab fragment of murine N1G9 antibody from theprimary immune response and of its complex with (4-hydroxy-3-nitrophenyl)a-

cetate. J Mol Biol 254: 208–222.82. Debler EW, Kaufmann GF, Kirchdoerfer RN, Mee JM, Janda KD, et al. (2007)

Crystal structures of a quorum-quenching antibody. J Mol Biol 368: 1392–1402.

83. Guddat LW, Shan L, Broomell C, Ramsland PA, Fan Z-C, et al. (2000) Thethree-dimensional structure of a complex of a murine Fab (NC10.14) with a

potent sweetener (NC174): An illustration of structural diversity in antigenrecognition by immunoglobulins. J Mol Biol 302: 853–872.

84. Monnet C, Bettsworth F, Stura EA, Le Du MH, Menez R, et al. (2002) Highlyspecific anti-estradiol antibodies: structural characterisation and binding

diversity. J Mol Biol 315: 699–712.

85. Corneillie TM, Fisher AJ, Meares CF (2003) Crystal structures of two complexesof the rare-earth-DOTA-binding antibody 2D12.5: ligand generality from a

chiral system. J Am Chem Soc 125: 15039–15048.86. Fleury D, Wharton SA, Skehel JJ, Knossow M, Bizebard T (1998) Antigen

distortion allows influenza virus to escape neutralization. Nat Struct Biol 5: 119–

123.87. Dokurno P, Bates PA, Band HA, Stewart LM, Lally JM, et al. (1998) Crystal

structure at 1.95 A resolution of the breast tumour-specific antibody SM3complexed with its peptide epitope reveals novel hypervariable loop recognition.

J Mol Biol 284: 713–728.88. Eigenbrot C, Meng YG, Krishnamurthy R, Lipari MT, Presta L, et al. (2003)

Structural insight into how an anti-idiotypic antibody against D3H44 (anti-tissue

factor antibody) restores normal coagulation. J Mol Biol 331: 433–446.89. Zahnd C, Spinelli S, Luginbuhl B, Amstutz P, Cambillau C, et al. (2004)

Directed in vitro evolution and crystallographic analysis of a peptide-bindingsingle chain antibody fragment (scFv) with low picomolar affinity. J Biol Chem

279: 18870–18877.

90. Lee JE, Kuehne A, Abelson DM, Fusco ML, Hart MK, et al. (2008) Complex ofa protective antibody with its Ebola virus GP peptide epitope: unusual features of

a V lambda x light chain. J Mol Biol 375: 202–216.91. Nogi T, Sangawa T, Tabata S, Nagae M, Tamura-Kawakami K, et al. (2008)

Novel affinity tag system using structurally defined antibody-tag interaction:application to single-step protein purification. Protein Sci 17: 2120–2126.

92. Diep Le KH, Lederer F, Golinelli-Pimpaneau B (2010) Structural evidence for

the functional importance of the heme domain mobility in flavocytochrome b2.J Mol Biol 400: 518–530.

93. Khan T, Salunke DM (2012) Structural elucidation of the mechanistic basis ofdegeneracy in the primary humoral response. J Immunol 188: 1819–1827.

94. Guddat LW, Shan L, Anchin JM, Linthicum DS, Edmundson AB (1994) Local

and transmitted conformational changes on complexation of an antisweetenerFab. J Mol Biol 236: 247–274.

95. Rietjens IM, Osman AM, Veeger C, Zakharieva O, Antony J, et al. (1996) Onthe role of the axial ligand in heme-based catalysis of the peroxidase and P450

type. Journal of Biological Inorganic Chemistry 1: 372–376.

96. Sundaramoorthy M, Terner J, Poulos TL (1995) The crystal structure ofchloroperoxidase: a heme peroxidase-cytochrome P450 functional hybrid.

Structure 3: 1367–1377.97. Hart-Davis J, Battioni P, Boucher JL, Mansuy D (1998) New catalytic properties

of iron porphyrins: Model systems for cytochrome P450-catalyzed dehydrationof aldoximes. J Am Chem Soc 120: 12524–12530.

98. Gajhede M, Schuller DJ, Henriksen A, Smith AT, Poulos TL (1997) Crystal

structure of horseradish peroxidase C at 2.15 A resolution. Nat Struct Biol 4:1032–1038.

99. Ricoux R, Girgenti E, Sauriat-Dorizon H, Blanchard D, Mahy JP (2002)Regioselective nitration of phenol induced by catalytic antibodies. J Protein

Chem 21: 473–477.

100. Ricoux R, Lukowska E, Pezzotti F, Mahy JP (2004) New activities of a catalyticantibody with a peroxidase activity: formation of Fe(II)-RNO complexes and

stereoselective oxidation of sulfides. Eur J Biochem 271: 1277–1283.

Crystal Structures of 13G10 and 14H7

PLOS ONE | www.plosone.org 18 December 2012 | Volume 7 | Issue 12 | e51128